基于粒子群算法的多目标跟踪优化传感器控制策略  

Multi-target tracking optimization sensor control strategy based on particle swarm optimization algorithm

作　　者：陈辉[1] 魏凤旗 赵永红 彭天曙 CHEN Hui;WEI Feng-qi;ZHAO Yong-hong;PENG Tian-shu(College of Electrical and Information Engineering,Lanzhou University of Technology,Lanzhou 730050,China;Gansu Province Changfeng Electronic Technology Co.LTD.,Lanzhou 730070,China;Gansu Provincial Computing Center,Lanzhou 730000,China)

机构地区：[1]兰州理工大学电气工程与信息工程学院,甘肃兰州730050 [2]甘肃长风电子科技有限责任公司,甘肃兰州730070 [3]甘肃省计算中心,甘肃兰州730000

出　　处：《兰州理工大学学报》2025年第2期88-93,共6页Journal of Lanzhou University of Technology

基　　金：国家自然科学基金(62163023,61873116,62366031,62363023);甘肃省基础研究创新群体(25JRRA058);中央引导地方科技发展资金项目(25ZYJA040);2024年度甘肃省重点人才项目(2024RCXM86);2023年度甘肃省军民融合发展专项资金。

摘　　要：针对多目标跟踪优化问题,提出一种基于粒子群算法的传感器控制策略.首先由泊松多伯努利混合(PMBM)滤波器的预测过程得到多目标预测状态,然后以此为先验信息通过粒子群算法以最大限度地接近各目标为准则求解传感器最优观测位置,并由传感器捕捉优质量测信息,最后由PMBM滤波器的更新过程得到优化多目标后验状态.仿真实验对比了多目标跟踪优化的效果,结果表明该传感器控制策略有更好的多目标跟踪精度.This paper presents a sensor control strategy based on particle swarm optimization for multi-target tracking optimization.Among the multi-target tracking methods,the Poisson multi Bernoulli mixture(PMBM)filter is widely used for its effective representation of undetected(existing but not detected)target information and more efficient recursive structure.First,the multi-target prediction state is obtained through the prediction process of the PMBM filter.Then,taking this as a priori information,the particle swarm optimization algorithm is employed to solve the optimal observation position of the sensor based on the criterion of maximizing the proximity to each target.The sensor then captures the optimal quality measurement information.Finally,the optimized multi-objective posterior state is obtained by the update process of the PMBM filter.Simulation experiments were conducted to compare the effectiveness of multi-target tracking optimization,and the results show that the proposed sensor control strategy has better multi-target tracking accuracy.

关 键 词：传感器控制 粒子群算法 多目标跟踪 泊松多伯努利混合 最优观测 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]